PROJECT SUMMARY Glioblastoma (GBM) is the most aggressive form of human cancers with very high fatality rate and short survival time, and the cancer cells aggressively infiltrate the brain and are intrinsically resistant to chemotherapy and radiation therapy. Intra-tumoral heterogeneity is a major challenge in therapeutic development for GBM patients because surgical acquisition of clinical specimens cannot be used to monitor the tumor progression and/or the underlying metabolic changes. Various neuroimaging methods have been used to study the morphology of the brain tumors. However, the need for noninvasively characterizing the brain tumors and their metabolic features has not been met, which should be critical for prognosis or for monitoring the tumor progression and response to treatment. It is well known that a common hallmark of the cancer cells is disrupted glucose metabolism, in which upregulated glycolysis is accompanied by inhibited mitochondrial oxidation, i.e., the ?Warburg effect?. Imaging the ?Warburg effect? and its spatial variability in brain tumors is a new attempt that can have a major impact on cancer research, particularly in the treatment of GBM, because therapies aimed at reversing the Warburg effect have shown promise in GBM ; however, great efforts are needed to develop novel metabolic imaging techniques to achieve the capabilities sought by clinicians. We have recently initiated a project aiming to develop a neuroimaging technique based on deuterium (2H) MRS (DMRS) detection of 2H-labeled brain metabolites following an administration of D-Glucose-6,6-d2 (d66). Our preliminary results indicate that the dynamic DMRS imaging can determine the cerebral metabolic rates of glucose (CMRGlc) and TCA cycle (VTCA), thus, the lactate production rate (CMRLac) in addition to the concentrations of deuterium-labeled glucose (Glc), mixed glutamate/glutamine (Glx) and lactate (Lac) in living brains. Furthermore, we demonstrated for the first time that the uncoupling between the glycolysis and oxidation in brain tumor can be quantitatively imaged via mapping the [Lac]/[Glx] ratio defined as an index of Warburg effect (IWE); and it has been shown that IWE is highly sensitive for distinguishing brain tumor from surrounding normal tissues. In this application, we are seeking NIH funding support to move forward with the DMRS imaging development through: i) integrated hardware and software development and the ultrahigh field MR technology to further boost signal-to-noise ratio (SNR), spectral resolution and spatiotemporal resolution; ii) testing the ultrahigh resolution DMRS imaging in healthy subject, and tumor patients and establishing a quantification model and imaging processing pipeline for future application; and iii) comparing the DMRS imaging results with the neuropathological and immunohistochemical findings of the biospecimens to understand the correlation between the DMRSI measurements and biological features of brain tumor. Our interdisciplinary research team with unique expertise is ready for a full-scale development of this highly innovative and cost-effective neuroimaging essential for basic research and clinic application in neuro-oncology.